Merge upstream AnkiDroid

.github/check_alpha_release.js

@@ -3,12 +3,21 @@
 const fs = require("node:fs");
 
 module.exports = async function checkAlphaRelease({ github, context, core }) {
+    async function skipRelease(reason) {
+        core.info(`Skipping scheduled release: ${reason}`);
+        await github.rest.actions.cancelWorkflowRun({
+            ...context.repo,
+            run_id: context.runId,
+        });
+        // Cancellation is asynchronous. Keep this step alive so cancelWorkflowRun can kill it.
+        await keepProcessAlive();
+    }
+
     const metadata = JSON.parse(
         fs.readFileSync("AnkiDroid/build/outputs/release-metadata.json", "utf8"),
     );
     if (metadata.releaseType !== "alpha") {
-        core.info("Skipping scheduled release: main is not on an alpha version.");
-        return false;
+        return await skipRelease("main is not on an alpha version.");
     }
 
     const releases = await github.paginate(github.rest.repos.listReleases, {
@@ -22,33 +31,34 @@ module.exports = async function checkAlphaRelease({ github, context, core }) {
         release => !release.draft && today - Math.floor(Date.parse(release.published_at) / day) < 7,
     );
     if (recentRelease) {
-        core.info(
-            `Skipping scheduled release: ${recentRelease.tag_name} was published fewer than seven UTC calendar days ago.`,
+        return await skipRelease(
+            `${recentRelease.tag_name} was published fewer than seven UTC calendar days ago.`,
         );
-        return false;
     }
 
     const latestAlpha = releases
         .filter(release => !release.draft && /^v\d+\.\d+\.\d+alpha\d+$/.test(release.tag_name))
         .sort((a, b) => Date.parse(b.published_at) - Date.parse(a.published_at))[0];
     if (!latestAlpha) {
-        core.info("Skipping scheduled release: no published alpha to compare against.");
-        return false;
+        return await skipRelease("no published alpha to compare against.");
     }
 
     const { data: comparison } = await github.rest.repos.compareCommitsWithBasehead({
         ...context.repo,
         basehead: `${latestAlpha.tag_name}...${context.sha}`,
     });
     if (comparison.status !== "ahead" || comparison.ahead_by === 0) {
-        core.info(
-            `Skipping scheduled release: no new commits descending from ${latestAlpha.tag_name}.`,
-        );
-        return false;
+        return await skipRelease(`no new commits descending from ${latestAlpha.tag_name}.`);
     }
 
     core.info(
         `Publishing an alpha: ${comparison.ahead_by} new commits since ${latestAlpha.tag_name}.`,
     );
     return true;
 };
+
+async function keepProcessAlive() {
+    // Fail after five minutes if cancellation never arrives.
+    await new Promise(resolve => setTimeout(resolve, 5 * 60 * 1000));
+    throw new Error("Timed out waiting for workflow cancellation.");
+}

.github/config.yml

@@ -5,12 +5,14 @@
 # Comment to be posted to on first time issues
 newIssueWelcomeComment: >
   Hello! 👋 Thanks for logging this issue. Please remember we are all volunteers here, so some patience may be required before we can get to the issue. Also remember that the fastest way to get resolution on an issue is to propose a change directly, https://github.com/ankidroid/Anki-Android/wiki/Contributing
+  Please read our [AI Policy](https://github.com/ankidroid/Anki-Android/blob/main/AI_POLICY.md) before contributing.
 
 # Configuration for new-pr-welcome - https://github.com/behaviorbot/new-pr-welcome
 
 # Comment to be posted to on PRs from first time contributors in your repository
 newPRWelcomeComment: >
   First PR! 🚀 We sincerely appreciate that you have taken the time to propose a change to AnkiDroid! Please have patience with us as we are all volunteers - we will get to this as soon as possible.
+  Please ensure your contribution meets our [AI Policy](https://github.com/ankidroid/Anki-Android/blob/main/AI_POLICY.md) before contributing.
 
 # Configuration for first-pr-merge - https://github.com/behaviorbot/first-pr-merge
 

.github/report_test_failures.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 # SPDX-License-Identifier: GPL-3.0-or-later
 
-"""Report failures and skipped tests from JUnit XML using the standard library."""
+"""Report failures, skipped tests, and overview summary from JUnit XML using the standard library."""
 
 import argparse
 import html
@@ -12,6 +12,19 @@
 import xml.etree.ElementTree as ET
 
 
+TABLE_HEADER = (
+    "## Test Results\n\n"
+    "| Suite | Result | Duration | Tests | Passed | Failed | Skipped |\n"
+    "| ----- | ------ | -------- | ----- | ------ | ------ | ------- |\n"
+)
+
+
+def format_duration(seconds):
+    minutes = int(seconds // 60)
+    rem_seconds = seconds % 60
+    return f"{minutes:02d}m {rem_seconds:06.3f}s"
+
+
 def read_ignored_tests(log_file):
     ignored = set()
     failed = set()
@@ -26,22 +39,61 @@ def read_ignored_tests(log_file):
     return ignored - failed
 
 
-def main(results_dir, log_file=None, pattern="TEST-*.xml", title="Test failures"):
+def get_report_duration(root):
+    suites = root.findall("testsuite")
+    if suites:
+        total = 0.0
+        for s in suites:
+            try:
+                total += float(s.get("time", 0) or 0)
+            except (ValueError, TypeError):
+                pass
+        if total > 0.0:
+            return total
+    try:
+        root_time = float(root.get("time", 0) or 0)
+        if root_time > 0.0:
+            return root_time
+    except (ValueError, TypeError):
+        pass
+    total = 0.0
+    for tc in root.iter("testcase"):
+        try:
+            total += float(tc.get("time", 0) or 0)
+        except (ValueError, TypeError):
+            pass
+    return total
+
+
+def main(results_dir, log_file=None, pattern="TEST-*.xml", title="Test failures", suite=None):
     failures = []
     skipped = []
     ignored = read_ignored_tests(log_file)
+    total_duration = 0.0
+    passed_count = 0
+    failed_count = 0
+    skipped_count = 0
+
     for report in sorted(results_dir.rglob(pattern)):
         try:
             root = ET.parse(report).getroot()
         except (ET.ParseError, OSError) as error:
-            failures.append(f"Could not read {report}: {error}")
+            if os.environ.get("GITHUB_ACTIONS") == "true":
+                print(f"::warning::Could not read {report}: {error}")
+            else:
+                print(f"Warning: Could not read {report}: {error}", file=sys.stderr)
             continue
 
+        total_duration += get_report_duration(root)
+
         for test in root.iter("testcase"):
+            name = f"{test.get('classname', '')}.{test.get('name', '')}"
+            has_failure = False
+            has_skipped = False
+
             for result in test:
                 if result.tag not in ("failure", "error", "skipped"):
                     continue
-                name = f"{test.get('classname', '')}.{test.get('name', '')}"
                 details = "".join(result.itertext()).strip()
                 message = result.get("message", "")
                 # Android can report assumption violations as failures or errors.
@@ -56,10 +108,30 @@ def main(results_dir, log_file=None, pattern="TEST-*.xml", title="Test failures"
                     "org.junit.AssumptionViolatedException",
                     "org.junit.internal.AssumptionViolatedException",
                 ):
+                    has_skipped = True
                     skipped.append(f"{name}\n{details or result.get('type', 'Test skipped')}")
                 else:
+                    has_failure = True
                     failures.append(f"{name}\n{details or result.get('type', 'Test failed')}")
 
+            if has_failure:
+                failed_count += 1
+            elif has_skipped:
+                skipped_count += 1
+            else:
+                passed_count += 1
+
+    total_tests = passed_count + failed_count + skipped_count
+    is_successful = failed_count == 0 and total_tests > 0
+    result_text = "SUCCESS" if is_successful else "FAILURE"
+    elapsed = format_duration(total_duration)
+
+    if suite:
+        print(
+            f"{suite}: {result_text} ({elapsed}) - "
+            f"{total_tests} tests: {passed_count} passed, {failed_count} failed, {skipped_count} skipped"
+        )
+
     for failure in failures:
         if os.environ.get("GITHUB_ACTIONS") == "true":
             # GitHub renders the escaped message as a multiline error in the log.
@@ -79,19 +151,27 @@ def main(results_dir, log_file=None, pattern="TEST-*.xml", title="Test failures"
         if os.environ.get("GITHUB_ACTIONS") == "true":
             print("::endgroup::")
 
-    if (failures or skipped) and os.environ.get("GITHUB_STEP_SUMMARY"):
-        with open(os.environ["GITHUB_STEP_SUMMARY"], "a", encoding="utf-8") as summary:
+    if os.environ.get("GITHUB_STEP_SUMMARY"):
+        with open(os.environ["GITHUB_STEP_SUMMARY"], "a+", encoding="utf-8") as summary:
+            summary.seek(0)
+            existing_content = summary.read()
+            if suite:
+                if "## Test Results" not in existing_content:
+                    summary.write(TABLE_HEADER)
+                summary.write(f"| {suite} | **{result_text}** | {elapsed} | {total_tests} | {passed_count} | {failed_count} | {skipped_count} |\n\n")
+
             if failures:
                 summary.write(f"### {title}\n\n")
                 for failure in failures:
                     summary.write(f"<pre>{html.escape(failure)}</pre>\n\n")
             if skipped:
+                skipped_title = f"Skipped tests ({len(skipped)})"
                 summary.write(f"<details><summary>{skipped_title}</summary>\n\n")
                 for skip in skipped:
                     summary.write(f"<pre>{html.escape(skip)}</pre>\n\n")
                 summary.write("</details>\n\n")
 
-    return 1 if failures else 0
+    return 1 if (failures or (suite and not is_successful)) else 0
 
 
 if __name__ == "__main__":
@@ -100,5 +180,6 @@ def main(results_dir, log_file=None, pattern="TEST-*.xml", title="Test failures"
     parser.add_argument("log_file", type=Path, nargs="?")
     parser.add_argument("--pattern", default="TEST-*.xml", help="Recursive report glob relative to results_dir")
     parser.add_argument("--title", default="Test failures", help="Failure heading in the job summary")
+    parser.add_argument("--suite", help="Suite name for markdown summary table")
     args = parser.parse_args()
-    sys.exit(main(args.results_dir, args.log_file, args.pattern, args.title))
+    sys.exit(main(args.results_dir, args.log_file, args.pattern, args.title, args.suite))

.github/with_emulator_cleanup.sh

@@ -0,0 +1,27 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+set -euo pipefail
+
+# Only for dedicated Linux CI runners with one emulator and -no-snapshot-save.
+# Forced shutdown must not interrupt saving a snapshot for subsequent jobs.
+stop_emulator() {
+  # #22613 `emu kill` can report success while QEMU keeps running and
+  # holds the emulator action's output pipes open. Bound both ADB and QEMU.
+  timeout -s KILL 5s "$ANDROID_HOME/platform-tools/adb" -s "emulator-$EMULATOR_PORT" emu kill || true
+
+  local emulator_pattern="^$ANDROID_HOME/emulator/qemu/"
+  local deadline=$((SECONDS + 20))
+  while pgrep -f "$emulator_pattern" > /dev/null; do
+    if ((SECONDS >= deadline)); then
+      echo "Emulator did not stop within 20 seconds; killing QEMU." >&2
+      pkill -KILL -f "$emulator_pattern" || true
+      break
+    fi
+    sleep 1
+  done
+}
+
+# Preserve the exit status
+trap stop_emulator EXIT
+"$@"

.github/workflows/label.yml

@@ -14,6 +14,7 @@ jobs:
     steps:
     - uses: actions/github-script@3a2844b7e9c422d3c10d287c895573f7108da1b3 # v9.0.0
       with:
+        retries: 3
         script: |
           let removeLabelsList = [
             "Pending Merge", 
@@ -42,6 +43,7 @@ jobs:
               owner: context.repo.owner,
               repo: context.repo.repo,
               issue_number: context.issue.number,
+              request: { retries: 0 },
               body: `Maintainers: Please [Sync Translations](https://github.com/ankidroid/Anki-Android/actions/workflows/sync_translations.yml) to produce a commit with only the automated changes from this PR.
 
           Read more about updating strings on the wiki,
@@ -80,6 +82,7 @@ jobs:
     steps:
       - uses: actions/github-script@3a2844b7e9c422d3c10d287c895573f7108da1b3 # v9.0.0
         with:
+          retries: 3
           script: |
             const I18N_FILES = [
                 "01-core",
@@ -133,6 +136,7 @@ jobs:
                 owner: context.repo.owner,
                 repo: context.repo.repo,
                 issue_number: context.issue.number,
+                request: { retries: 0 },
                 body: `> [!IMPORTANT]
             > **Maintainers**: This PR contains https://github.com/ankidroid/Anki-Android/labels/Strings changes
 
@@ -207,6 +211,7 @@ jobs:
     steps:
       - uses: actions/github-script@3a2844b7e9c422d3c10d287c895573f7108da1b3 # v9.0.0
         with:
+          retries: 3
           script: |
             const creator = context.payload.sender.login
             const opts = github.rest.issues.listForRepo.endpoint.merge({
@@ -240,6 +245,7 @@ jobs:
     steps:
       - uses: actions/github-script@3a2844b7e9c422d3c10d287c895573f7108da1b3 # v9.0.0
         with:
+          retries: 3
           script: |
             await github.rest.issues.addLabels({
               issue_number: context.issue.number,

.github/workflows/merge_requirements.yml

@@ -9,9 +9,7 @@ on:
 
 permissions: {}
 
-concurrency:
-  group: ${{ github.workflow }}-${{ github.ref }}
-  cancel-in-progress: true
+# Don't cancel runs: this is a fast check which shows as a failing check when canceled.
 
 jobs:
   squashMerge:

.github/workflows/publish.yml

@@ -27,6 +27,9 @@ concurrency:
 jobs:
   check_release:
     name: 'Check whether a release is due'
+    permissions:
+      contents: read
+      actions: write # Cancel this run when a scheduled publish is skipped.
     if: github.event_name == 'schedule' && github.repository == 'ankidroid/Anki-Android' && github.ref == 'refs/heads/main'
     timeout-minutes: 20
     runs-on: ubuntu-latest

.github/workflows/screenshot_comment.yml

@@ -64,7 +64,7 @@ jobs:
         name: Check if there are valid files
         shell: bash
         run: |
-          # Find Roborazzi diff PNGs in the downloaded artifact (in <TestClass>/).
+          # Find Roborazzi diff PNGs in the downloaded artifact (in <module>/<TestClass>/).
           mapfile -t files_to_add < <(find . -name "*_compare.png" -type f)
           if [ ${#files_to_add[@]} -gt 0 ]; then
             exist_valid_files="true"
@@ -82,13 +82,14 @@ jobs:
           BASE_REF: ${{ github.event.workflow_run.event == 'pull_request' && github.event.workflow_run.pull_requests[0].base.ref || github.event.repository.default_branch }}
         shell: bash
         run: |
-          # Find Roborazzi diff PNGs to mention in the comment (in <TestClass>/).
+          # Find Roborazzi diff PNGs to mention in the comment (in <module>/<TestClass>/).
           mapfile -t files < <(find . -name "*_compare.png" -type f | sort)
           total=${#files[@]}
           # Cap the full list so we remain within GitHub's comment limit
           max_details=100
 
-          # Group by test class - the file's immediate parent directory.
+          # Group by test class, module prefix is dropped for readability.
+          # Reconsider it once we are heavily mutlimodule.
           declare -A class_counts class_files
           for file in "${files[@]}"; do
             class_dir="${file%/*}"